Most research suggests that alcohol use peaks in the college years then declines into the mid-thirties; understanding this maturing out process may clarify who is at risk for ongoing alcohol-related problems (Jochman & Fromme, 2010). However, there is evidence that some individuals mature out much earlier: downward drinking trends for some individuals begin early in the college years, with as many as one third of college students decreasing their drinking (Baer et al., 2001). Work using the developmentally sensitive method of trajectory analysis, which helps identify different subgroups of a population who experience different developmental trends, has consistently identified a group that decreases drinking beginning during the first year of college: an early maturing out group (e.g. Schulenberg et al., 1996; Greenbaum et al., 2005). It is crucial for risk research, and thus for the public health, to identify factors that differentiate those who decrease their drinking early from those who persist in high levels of alcohol consumption; doing so would clarify risk for college-related alcohol problems and alcohol use disorder, and aid in developing earlier targeted prevention and intervention efforts. Past research has examined contextual and individual difference factors in relation to varying trajectories (Chassin et al., 2002; Gates et al., 2016; Jackson & Sher, 2005). However, those studies did not address early maturing out. The proposed research emphasizes two possibilities: 1) perhaps those who mature out early are facing adult-like responsibilities such as paying for their educations/supporting their families and/or 2) perhaps those who persist have higher levels of personality and learning-based risk factors compared to those who mature out early. I propose to test both possibilities. Concerning the latter, I apply a new model that integrates the personality trait of negative urgency, the tendency to act rashly when in a highly negative mood, with psychosocial learning in the form of alcohol expectancies, which has proven quite effective in understanding drinking risk (Peterson et al., 2018). The research component of this proposal involves using a four-wave longitudinal design to (a) identify a group that decreases drinking levels early in college and a group that persists using trajectory analyses and (b) test whether these groups can be differentiated on the basis of contextual or individual difference factors. Studying factors that influence early maturing out is novel. A successful test would have clear clinical and public health implications, as the proposed project is likely to provide direction for targeted intervention during the college years. The training component of this proposal includes training in the following: 1) the execution and management of a longitudinal study, including participant recruitment, data collection, and follow-up; 2) complex statistical analyses, such as multilevel modeling; 3) the emerging adulthood developmental literature, which is likely to prove helpful in understanding early maturing out; 4) further advanced training in research ethics; 5) preparation of research reports for publication; 6) mentoring of junior graduate and undergraduate students.